Abstract
Myocardial rupture is often considered a complication after an acute myocardial infarction (MI), and it occurs quite frequently in patients. Without an autopsy or imaging evidence, sudden cardiac death at the time of acute MI can easily be attributed to one of a variety of other reasons, including muscle death, intractable arrhythmia, heart block, or pulmonary embolism, with the diagnosis of rupture often forgotten or least prioritized [1]. According to Harvard Medical Research, the average age for Myocardial Infarction in men of the United States is 65. That is also the reason why coronary artery diseases, including MI, cardiac arrest, and heart failure, have been labeled a disease of senior citizens. On the other hand, the initial presence of acute MI corresponds to the acute ruptures' occurrence. As per the "Valsartan in Acute Myocardial Infarction" (VALIANT) trial conducted on some 14,703 patients with either clinical congestive heart failure or reduced ejection fraction of <; 40% within ten days of an acute MI, provides some apprehension of the timing of death from myocardial rupture later after the MI [1]. Machine learning attempts to make the prediction of a Myocardial Rupture after Myocardial Infarction more accurate. After employing the Random Forest model from the machine learning algorithm, we discovered the feature importance, feature performance, and most important factor in the prediction. Features such as age, gender, the number of myocardial infarctions in the anamnesis, exertional angina pectoris in the anamnesis, Functional Class (FC) of angina pectoris in the last year, hypertension, and chronic heart disease were judged to reach a substantial outcome. In the following research, we have attempted to describe the problem statement and how it can be resolved using the machine learning algorithm.